Disparities and similarities in the spatiotemporal dynamics of flash and slow droughts in China

Abstract Climate warming has induced significant transitions from slowly-developing droughts to rapidly-developing flash droughts in China, causing broad impacts on ecosystems, hydrological regimes, and society. To date, most studies focused on temporal evolution of flash droughts, while neglected the spatial expansion which is essential for understanding their origins and spatial propagations, especially for mega flash droughts. Based on the long-term (1940–2022) dataset of the 5th generation of the European ReAnalysis, here we use a three-dimensional drought identification method to analyze the disparities and similarities in the spatiotemporal dynamics of flash and slow droughts at the subseasonal time scale over China. Although half of the flash and slow droughts are characterized by small areas (<5000 km 2 ), short durations (30–45 d) and short propagation distances of drought centroids (<50 km), the probability of large-scale (>30 000 km 2 ) flash droughts with long propagation distances (>100 km) is twice of slow droughts. Moreover, global and local spatial autocorrelation analyses reveal that South China (SC) and North China are hotspots for large-scale flash and slow droughts, respectively, and they both show significant increasing trends (0.11–0.12 events/decade) during 1940–2022. Without these large-scale droughts, there is no obvious difference in spatial distributions of the frequency of flash and slow droughts. Despite disparities, both large-scale flash and slow droughts show a preferential westward propagation, with 60%–67% of the movements consistent with the pathways of atmospheric water vapor flux anomaly. Our study urges the understanding and prevention of large-scale flash drought events, especially in SC.